Constraints for high precision polarization detection at ...

Alessia RitaccoApril 15th 2021

Constraints for high precision polarization detection at mm and sub-mm wavelengths from ground and space

Alessia RitaccoLaboratoire de Physique de l’École normale supérieure (LPENS)Institut d’Astrophysique Spatiale (IAS) Paris


Outline

Technological development to unveil:

● Magnetic fields physics NIKA2 from IRAM 30m telescope

● Cosmic Microwave Background polarization B-modes CMB experiments - ex. LiteBIRD satellite

Polarization detection constraints:

● Mitigating systematics effects● Absolute calibration● Astrophysical foregrounds

2


Science: - Star formation;- Galactic and extragalactic physics;- Cosmology through the SZ effect in galaxy clusters;- Solar system.

NIKA2 at IRAM 30m telescope

30 m

Telescope rotates on an alt-azimuth mechanism with a tracking precision of < 3 arcsec

NIKA2 continuum camera: 6.5 arcmin of FoV

2mm band: 125-170 GHz FWHM 17 arcsec

1mm band: 240-280 GHz FWHM 11 arcsec

The observatory operates 24 hours a day every day of the year

NIK

A2

2mm

ba

nd

NIK

A2

1mm

ba

nd

3

Perotto et al. 2020


Motivations for NIKA2 polarimeter development:magnetic field structures in galactic regions

Need of high angular resolution observations to resolve the width of filaments ~ 0.01-0.05 pc

Herschel satellite results suggest a filamentary paradigm of star formation

Large scale MHD turbulent flows generate filaments

Gravity fragments the densest filaments into prestellar cores

Polaris - Herschel/Spire 250 μm Taurus B211/3 - Herschel

Taurus: columns density + B lines

Planck int. results XXXV

Planck polarization results reveal a well organized magnetic field at large angular scales

Ref: Protostars and Planets VI review André, Di Francesco, Ward-Thompson+2014

4


Technological challenges, systematics and calibration

?

What’s next

Cosmology

5


Beyond Planck satellite observations

Credits: ESA and Planck collaboration

CMB Temperature CMB Polarization

Planck satellite provided the best full-sky maps of Cosmic Microwave Background (CMB) to date in both temperature and polarization.

CMB polarization patterns can be expressed as the superposition of two different modes: the E-modes, of even parity, and the B-modes.

B-modes can only be produced by primordial gravitational waves in the early universe.If detected they will probe the existence of an inflationary epoch and give us access to a physics beyond the current Standard Model.

6


TT spectrum: cosmological parameters from density perturbations

EE spectrum: model coherence, break degeneracies

BB lensing spectrum: gravitational lensing of EE-modes, large-scale structures

BB primordial spectrum: tensor perturbations from primordial GW background, scaled by tensor to scalar ratio r

Best upper limit is r < 0.044 [Planck low-l, BICEP2/Keck and Planck dust, Tristam + 2020]

Experiments under development are designed to target σ(r) <10-3

LiteBIRD 2 ≤ l ≤ 300 CMB-S4 30 ≤ l ≤ 5000

CM

B-s4

+2

016

LiteBIRD

CMB-S4

Scientific motivations: CMB-B modes detection constraints

7


CMB experiments panorama

The next generation "Stage-4" ground-based CMB experiment.Current Stage IIISouth Pole Telescope:~ 16400 detectorsfreq. 95-150-220 GHzBICEP3:2560 detectors @ 95GHzKeck Array telescopes:~29000 detectors @ 35, 95, 150, 220, 270 GHz

GROUND

SPACELiteBIRD satellite conceptual design

Uncertainty expected on tensor-to-scalar ratio r < 10-3

Noise challenge:→ decreases increasing the number of detectors

8


High precision polarization detection challenges

Instrumental improvements: Increasing SNR and reducing instrumental noise

- Arrays of high sensitive detectors

- Half wave plates:

from ground it also dramatically reduces correlated atmospheric

noise.

Calibration: high control of systematics, absolute accuracy of the polarization reconstruction.

Astrophysical foregrounds:Component separation → large frequency range coverage

9


NIKA2 polarimeter Half wave plate modulator

Two 260 GHz arrays measure the two linear polarization components

Modulating the polarization signal with a rotating half wave plate has numerous advantages:

● It separates the polarization signal from the unpolarized one

● A single polarized detector measures simultaneously the polarization parameters Stokes Q and U

● The signal is shifted far from 1/f noise

● It mitigates several systematics (atmosphere, thermal drifts, beam mismatch, asymmetries...)

● But it can introduce new systematic effects (Ritacco+17, D’Alessandro+2019) that need to be carefully addressed

10


Data model

11


Control of Half Wave Plate systematics

Instrumental polarization:

Linear polarization from unpolarized sky emissionHWP cross-polarization → Leakage of polarisation from one axis to the otherDrifts of the HWP temperature with time → Different emissivitiesDeterioration of the beam passing through the HWP (ellipticity) Reflections at non-normal incidence can be detected at 4f (I > Q /U)…

Data modelIncoming polarization modulated at 4 times the HWP rotation

frequencyAtmospheric absorption

HWP Synchronous Signal

Atmospheric emissionDetector

noise

Glitches terms

Electronic noiseCryogenic

noiseInstrumental polarization

12

Ritacco et al. A&A 599, A34 (2017)


Mitigating the systematic effects* Absolute flux calibration instabilities due to atmospheric fluctuations* Polarization efficiency* Level of instrumental polarization → I to P leakage * Absolute angle calibration* Side lobes …

13


Mitigating the systematic effects

NIKA-example: continuous rotating multi-layers HWP produced a modulation also of the background signal

This HWP synchronous signal (HWPSS) is peaked at harmonics of the HWP rotation frequency 𝜈 Ritacco et al. 2017 modeled the HWPSS (and subtracted it per TOI)

Ritacco et al. A&A 599, A34 (2017)

14

* Absolute flux calibration instabilities due to atmospheric fluctuations* Polarization efficiency* Level of instrumental polarization → I to P leakage * Absolute angle calibration* Side lobes …



NIKA/IRAM 30m case addressed inRitacco et al. A&A 599, A34 (2017)

Stokes Q Stokes UIt depends on the optical elements including the HWP + detectors

Stokes I

Unpolarized source: we expect NO signal in Stokes Q and U maps

Stokes I

15




NIKA/IRAM 30m case addressed inRitacco et al. A&A 599, A34 (2017)

It can be modelled and subtracted if it is stable and common to all the detectors

Stokes Qafter correction

Stokes Uafter correction

IP residual: 0.6 % of total intensity→ to be accounted in the overall polarization uncertainties budget

16




NIKA2/IRAM 30m case preliminary investigation discussed inAjeddig et al. EPJ Web of Conferences 228, 00002 (2020)

NIKA2 has presented a more complicated pattern, which depends on many parameters

- Leakage effect not stable → difficult to be modelled- A dependency with source elevation and focus hasbeen observed and now we are close to model it and subtract it to all data sets

17



Mitigating the systematic effects* Absolute flux calibration instabilities due to atmospheric fluctuations* Polarization efficiency* Level of instrumental polarization → I to P leakage * Absolute angle calibration* Side lobes …

18


A miscalibration of the absolute polarization angle by ∆ψGal

will lead to a mixing of E and B modes. In the CMB and because C

lEE >> C

lBB, this is often referred to as an “E to B leakage” and reads

Spurious bias component

Accuracy in the calibration of the polarization angle:- Ground calibration: very good but need to be validated during operations

- External calibration source: good accuracy but never done before, instrumental limits ?!

- Self-calibration: we expect no scientific signal from TB and EB → only instrumental

→ Losing constraints on fundamental phenomena

- Sky calibration: frequency dependence, time variability → Best option: CRAB NEBULA

Challenge for CMB-B modes detection constraints

19


Accuracy of the polarization detectionA sky calibrator: the Crab nebula

* Absolute angle calibration

Aum

ont+

10

Rita

cco+

18

Rita

cco+

20

XPOL/IRAM 30m NIKA/IRAM 30m NIKA2/IRAM 30m

The Crab Nebula (Tau A) is a plerion-type supernova remnant, observed from radio to X-rays

The microwave emission has an extension of about 5′ × 7′

Most intense polarized source in the microwave sky, at angular scales of few arcminutes

Highly polarized synchrotron emission with a polarization fraction of ~ 20%

It is relatively isolated in the microwave sky within 1 degree scale

Freq. 89 GHz FWHM 27′’

Freq. 150 GHz FWHM 17′’

Freq. 260 GHz FWHM 11.2′’

20


⋆Planck HFI fluxes have been recomputed here by using aperture photometry techniques

Spectral energy distribution (SED)

Polarized intensityflux [Jy]

Intensityflux [Jy]

The polarization spectral index is consistent with the total power index confirming that the synchrotron radiation from a single population of relativistic electrons is responsible for the emission of the nebula.𝜷= −0.323 ± 0.001 ; 𝜷

POL = −0.347 ± 0.026.

Ritacco et al. 2018, A&A, 616, A35

21


𝝍 = ½*atan(U/Q) p = sqrt(Q2+U2)/I

Pol. angle Pol. degreeNIKA/IRAM 30m

XPOL/IRAM 30m

A variation from small to large angular scales is observed on both the polarization angle and degree

The polarization direction appears stable with the frequency and constant within a radius of 2 arcmin from the Nebula center

In order to compare with CMB experiments results let’s check the integrated flux across the source

Ground based high angular resolution observations

Aum

ont+

10R

itacc

o+18

22


Excluding Polka data outliers

Constraints of the absolute calibration

* Compilation of: WMAP [Weiland+11] Planck-LFI [Planck 2015 XXVI], Planck-HFI, re-analyzed in [Ritacco+18]) XPOL\IRAM-30m [Aumont+10] and NIKA\IRAM-30m [Ritacco+18]

Total weighted polarization angle average:

𝝍 = -88.26°± 0.27°

J. Aumont , J.F. Macías-Pérez, A. Ritacco, N. Ponthieu,

A. Mangilli A&A 634, A100 (2020)

23


Combining current (and future) measurements

* Assumption of constant polarization angle is necessary to lower the D

lBB bias

* Under the crucial assumption that the polarization angle is constant with frequency, the combined error 𝝈(𝜓) ~ 0.27° could allow to probe r = 10-2

* Future accurate measurements of the Crab are needed to meet the requirements of future CMB experiments (𝝈(𝜓) ~ 0.1° ) to measure r = 10-3

- NIKA2Pol high sensitive maps at 260 GHz under investigation

- SCUBA2Pol at 353 GHz proposal submitted

Power spectrum bias from E−B mixing due to the miscalibration of the absolute polarization angle Aumont+2020

24


In total intensity the large angular scale spatial variation of the dust spectral energy distribution (SED) is analytically described by introducing two additive SED corrections factors. The dust model is thus:

where f(𝛎) is the Planck modified black-body function.

Foregrounds challenge

Main polarized foregrounds:

● Synchrotron emission

● Dust emission

Delouis et al. 2021

The correlation of these results with polarization data is under study (Ritacco+2021, in prep) and will be crucial to understand the behaviour of the dust polarization and prepare data analysis modelling for e.g. CMB-S4, LiteBIRD.

25


Summary

★ High control of the systematics induced by optical elements;

★ Modelling optical effects and propagating them into data analysis is the only way to choose the best configuration of polarization modulators;

★ Sky absolute calibration in a large frequency range is crucial for next generation of CMB experiments○ current measurements could allow to probe r = 10-2;

○ future accurate measurements of the Crab (e.g. NIKA2, SCUBA-2) are needed to meet the requirements of future CMB experiments to measure r = 10-3 (e.g. LiteBIRD, CMB-S4).

★ Spatial dust SED variation has to be carefully accounted for in the component separation data modelling.

26

Alessia RitaccoApril 15th 2021 27

BACKUP SLIDES


Kinetic Inductance Detectors

KIDs are RLC superconducting resonators

KIDs are intrinsically suited for Frequency Domain Multiplexing

Sensitive to one polarization

Sensitive to both linear polarizations

NIKA/NIKA2 detectors

Hilbert geometry

28


NIKA2 KID array

● 2 mm array: 616 pixels 4 feedlines● 1.15 mm arrays: 1140 pixels 8 feedlines

29

Constraints for high precision polarization detection at ...

Documents

Transcript of Constraints for high precision polarization detection at ...